DESCRIPTION (provided by investigator): There is growing evidence that some air pollution constituents perturb fetal development. Some studies point to motor vehicle exhaust pollutants specifically as causative agents of interest. However, to date almost all existing studies have focused on criteria air pollutants (CO, NOx, particles, SO2) rather than specific air toxics and have used data from existing government air monitors to assess exposure. These air monitoring data are unlikely to adequately capture the effects of primary motor vehicle exhaust pollutants whose concentrations vary based on proximity to sources. Land use-based regression (LUR) models are an attractive alternative to assess exhaust toxin exposures, since they can be applied to large populations and account for neighborhood- scale variations in pollutant concentrations. Thus, the specific aims of this study are to: (1) examine associations between prenatal exposure to motor vehicle air toxics and low birth weight (LBW) and preterm birth in women residing in Los Angeles County, California between 1994-2006 using both ambient measurement data and LUR models;and (2) gain information about how LUR models built on NOx measurements reflect exposures to specific toxins thought to have biological relevance for these outcomes. Specifically, using existing data sources, we will examine associations between the exhaust toxins polycyclic aromatic hydrocarbons (PAHs), benzene, and metals and adverse birth outcomes in an epidemiologic cohort study. We will assess pregnancy-specific exposures to these compounds using a unique resource of air toxics monitoring data from existing government networks coupled with special measurement studies recently conducted in LA. We will also apply LUR models we are currently developing for the LA Basin based on NOx measurements (as a tracer for vehicle exhaust) to obtain more spatially refined exposure metrics. Using data from a nested case-control sample of LA County births we previously collected in 2003, we will also examine confounding of air pollution effect estimates by a large number of risk factors for which information is not available on birth certificates. These data will also allow us to examine exposure misclassification due to mobility during pregnancy. The proposed study will contribute significantly to understanding the spatial patterns of motor vehicle toxin exposure among pregnant women in LA County, a region heavily polluted by vehicular emissions. Information will be generated on the association between NOx-based traffic pollutant models, increasingly used in health effects assessment, and measured air toxics concentrations. Most important, we expect the findings to enhance our understanding of whether motor vehicle air toxics are the primary drivers of associations between fetal development and criteria air pollutants (CO and particles) observed in previous studies. If the proposed pilot study is able to pinpoint specific toxins of greater relevance to birth outcomes, future research using personal and biomonitoring would be able to target these compounds. PUBLIC HEALTH RELEVANCE Perturbations of fetal development, including preterm and low weight birth, elevate the risk of mortality and morbidity in childhood, and may also impact health in adulthood increasing the risk for cardiovascular disease and diabetes. The expected growth in motor vehicle usage in both developed and developing countries underscore the public health relevance of traffic sources of air pollution and their impact on fetal development.